Although people with disabilities represent 15% of the total US population (1), research on the effect of the built environment on physical activity among people with disabilities is limited (2). However, evidence suggests that the aspects of the built environment that encourage physical activity among the general population may facilitate physical activity in disabled populations (3). Few studies have assessed the effect of the built environment on physical activity behaviors of people with disabilities. These studies have primarily analyzed accessibility to recreation programs and fitness facilities (2,4), environmental supports that affect physical activity (3,5), and community mobility as influenced by the built environment (6). The purposes of this study are to investigate whether the perceptions of people with disabilities about characteristics of the built environment are related to their physical activity behaviors and whether this relationship differs from that of people without disabilities. Our hypothesis is that accessibility barriers in the built environment give people with disabilities less positive perceptions of the built environment that cause them to engage in less physical activity than people without disabilities.

Extensive epidemiologic studies have demonstrated strong associations between physical activity and health. Physical activity reduces the risk of many of the major causes of illness or death in the United States, such as cardiovascular disease, high blood pressure, obesity, and depression (7). Many studies have concluded that well-designed built environments encourage physical activity. Heath et al (8) identified 13 cross-sectional studies published between 1993 and 2003 that support community-scale urban design and land-use policies and practices that increase physical activity, such as walking and biking infrastructure, proximity to recreation areas, and the aesthetic and safety aspects of the street-scale built environment. More recent research indicates that the following factors affect health-related physical activity: urban form (9-11), neighborhood design (12,13), neighborhood environmental quality (14), street type (15), vegetation (16), the proportion of green space, residential density, and the general impression of activity-friendliness of the neighborhood (17). Because physical activity is significantly affected by the built environment, "how we design the built environment may hold tremendous potential for addressing many of the nation's greatest current public health concerns" (18).

The design of the built environment disproportionately affects people with disabilities in comparison with their peers (19,20). Conditions in the built environment may create barriers to people with disabilities that reduce opportunities to engage in physical activity (2,5,21,22) and contribute to disparities that persist in nearly every aspect of health among people with disabilities (23). In addition to improperly implemented regulations of the Americans with Disabilities Act, features such as pathway texture, disconnected pedestrian ways, signage, and slope have more influence on the participation of people with disabilities in physical activity than on people without disabilities (5). Healthy People 2010 recognizes this health disparity and reports that 56% of adults with disabilities do not engage in leisure-time physical activity (LTPA) compared with 36% of people without disabilities (24). The prevalence of sedentary behaviors among people with disabilities increases their susceptibility to chronic diseases and secondary health conditions (25). Thus, Healthy People 2010 includes a developmental objective to increase LTPA participation among people with disabilities by reducing environmental barriers (24).

For the total study sample who met the inclusion criteria (N = 4,947), the mean age of participants was 42 years; 62% were female, 91% had graduated from high school, and 44% reported an annual income between $25,000 and $75,000 (Table 1). The total sample was racially and ethnically diverse; 58% were white and 30% were Hispanic.

Seventeen percent of respondents reported disabilities (n = 849). Of these, 60% reported LTPA outside of work (compared with 78% of those without disabilities) and 43% met recommended levels of physical activity (compared with 54% of respondents without disabilities). In general, people with disabilities perceived their neighborhoods less favorably and reported less use of community facilities for physical activity. To all questions regarding community facilities, people with disabilities more frequently reported that such facilities did not exist in their community (Table 1).

Both people with and without disabilities who reported using private recreation facilities (odds ratios [OR], 2.04 and 3.07, respectively) and trails, parks, playgrounds, or sports fields (OR, 2.12 and 2.36, respectively) were significantly more likely to report LTPA than those who reported no such use (Table 2). People with disabilities who described the street lighting for walking at night in their neighborhood as "very good" or "good" were less likely to participate in LTPA than those who described street lighting as "very poor" (OR, 0.34 and 0.50, respectively). However, the perceptions of people without disabilities about the condition of nighttime street lights proved unrelated to their LTPA. In contrast, the LTPA of people without disabilities was strongly associated with perceptions of neighborhood safety when participants classified their neighborhood as "extremely safe" (OR, 2.82), "quite safe" (OR, 3.05), or "slightly safe" (OR, 1.70) compared with "not at all safe" (Table 2).

For people with disabilities, the reported presence of neighborhood sidewalks, when compared with reported absence of sidewalks, was associated with meeting recommended PA levels (OR, 1.59) (Table 3). People with disabilities who reported their neighborhoods were "not very pleasant" places to walk (OR, 4.04) were more likely to meet physical activity recommendations than were people with disabilities responding that their neighborhoods were "not at all pleasant" places to walk. Furthermore, people with disabilities who reported their neighborhoods were "slightly safe" from crime, compared with "not at all safe," were less likely to meet physical activity recommendations (OR, 0.36). For people without disabilities, the use of private recreation facilities (OR, 1.47) and trails, parks, playgrounds, or sports fields (OR, 1.20) for physical activity was significantly related to meeting recommended levels of physical activity.

The use of walking trails, parks, playgrounds, or sports fields was significantly associated with the LTPA of people with disabilities. This study supports previous studies reporting that access to trails, parks, and playgrounds is associated with LTPA (2,3). Furthermore, a much greater proportion of people with disabilities reported that their community does not have these facilities or that they do not use them than did people without disabilities. This is probably the result of the facilities being inaccessible to people with disabilities, resulting in an unmet demand for access to trails, parks, playgrounds, and sports fields.

Among people with disabilities who reported LTPA, street lighting for walking at night was one of the neighborhood characteristics significantly related to reported LTPA, although the data indicate a negative association. No association was found between nighttime lighting and meeting recommended levels of physical activity for this demographic group. Furthermore, because the question about adequate street lighting specifically mentions walking, the results may have been different if other physical activities were examined. Although this finding is subject to confounding factors such as the definition of LTPA (eg, access to a car, daytime versus nighttime workout preferences, sidewalk conditions), it indicates that lighting is an important component of accessibility. People with disabilities using lighted areas at night are not using them as venues for physical activity but rather as nighttime accessible pathways.

The presence of neighborhood sidewalks is associated with meeting recommended levels of physical activity for people with disabilities. This finding coincides with the other issues already discussed related to accessibility. The presence and connectivity of sidewalks within a neighborhood not only provide venues for exercise but may also be the primary means of access to other facilities conducive to physical activity (17). Although parks, boulevards, boardwalks, and fitness facilities are still accessible to people without disabilities when sidewalks are lacking or poorly connected, people with disabilities are disproportionably affected (13,17).

LTPA was strongly associated with perception of neighborhood safety for people without disabilities, but people with disabilities did not show the same association. Yet, people with disabilities were more likely than people without disabilities to indicate that their neighborhood was unsafe. Chronic exposure to inaccessible and unsafe neighborhood environments may have conditioned people with disabilities to be less influenced by such conditions.

People with disabilities were less likely to participate in LTPA and meet recommendations for physical activity than people without disabilities (Table 1). These findings are consistent with those of other studies reporting inactivity prevalence among populations with disabilities (8), but actual LTPA and physical activity levels may differ from those reported here because of confounding factors such as self-reported activity, inaccuracies in memory recall, and the use of physical activity questions that may not adequately measure this demographic group (2,8). People with disabilities also were less likely to use any of the reported community facilities for physical activity except shopping malls, which are typically adequately accessible environments. In all instances, people with disabilities were more likely to report that the analyzed facilities did not exist in their community. This finding may indicate that people with disabilities are living in more dilapidated areas but more likely signifies a lack of accessibility and thus a functional unavailability of facilities (4).

As acknowledged by the Centers for Disease Control and Prevention, the use of BRFSS data in this circumstance creates some limitations. All information used in this study was gathered from surveys with a response rate below 50%. Further, the subjectivity of terms such as "pleasantness" and "safe" is vulnerable to respondent interpretation. However, the survey questions were designed to assess respondents' perceptions of their neighborhood. Studies to examine the validity of these perceptions in relation to actual environmental characteristics could further enhance the findings of this study. Likewise, findings might have differed greatly, as suggested elsewhere (2,8) if different questions were used that more adequately represented people with physical disabilities and more specifically analyzed their typical physical activities. For example, the Physical Activity and Disability Survey developed by Rimmer, Riley, and Rubin (30) allows people with disabilities more flexibility in describing their physical activity participation by acknowledging that commonly surveyed activities (eg, walking, bike riding, using private recreation facilities) might not accurately represent this population's physical activities. This study found significant differences between the LTPA and physical activity of people with and without disabilities. Results also indicate that the built environmental has similar effects on people with and without disabilities.